Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
×

Comment Re:The poison pin ... (Score 1) 340

Somewhere else, maybe... at the border crossing they have near infinite power to mess with you by insisting on an extended identity, security and luggage check and usually to detain you for a short while too for almost no pretext at all. In fact your "defective phone" is now a possible terrorist bomb, let's just put you in a holding cell until we can determine it's not.

Comment Re:Ah, come one, don't we trust the Feds? (Score 1, Troll) 90

Some of us have worked on the ISP side of the house (disclosure: I worked for a small one that was crushed by Time Warner a long time ago) and view the Netflix debacle in a different light. Netflix has a history of trying to pass their costs onto third parties, by abusing settlement free peering, pushing their "Open Connect" devices on ISPs without offering to pay the usual co-location expenses, or trying to cheap out on envelopes that wound up jamming in sorting machines and causing USPS all manner of difficulties. That one turned into a major spat as I recall, with USPS having to threaten to revoke their bulk mailing/pre-sort price discounts before Netflix was willing to back down.

The long standing model for internet traffic has been sender pays. If you're dumping more traffic into my network than you take off my hands you pay me to get it closer to its destination. If you're taking more off my hands than I'm taking from you then I pay you. In the final example, we exchange roughly equal amounts of traffic and agree to do so without remuneration.

Is that model still valid today? It's hard to say. It did build the internet as we know it today, for better or worse. It would be easier for me to be sympathetic if this wasn't a pissing contest between Netflix and ISPs. The arrogance of Netflix is truly astounding, from my perspective as someone who worked in the ISP business, and I see it as billionaires arguing with other billionaires about who should foot the bill for their respective business models.

Comment Re:Interpreting these conditions (Score 1) 188

The controversial part, as I understand it, is the difference in interpretation of a license's conditions. For example, the difference between an "aggregation" and a "combined work" in the GPLv2 confused at least one Slashdot user.

Actually the ugliest part of the GPL which is clear as ink in law is what - if anything - makes inter-module communication derivative. The theory of derivative works mainly involve sections or elements reappearing in the derivative, like a composite made from a photo. It doesn't cover interfaces where independently developed code calls each other at all. If I wrap a GPL library into a web service, is calling it derivative? If the answer is yes, the GPL is extremely viral. If the answer is no, the GPL is in big trouble. Which is why you never get a straight answer.

This directly links in with the "mere aggregation" clause, if you can for example distribute a distro that has an application that sends mail and a mail server without those being derivative, can you also distribute proprietary software and this web service? Your software needs it, this software happens to provide it but it could in theory be provided by a different implementation. I'm sure Stallman says no, but it's entirely unclear to me if a judge would agree.

Comment Re:How does stingray connect to the wider network? (Score 1) 90

Which may have value in intelligence operations aboard but is completely pointless domestically, where the law requires that telecommunications providers provide for lawful interception, interception that can happen while you sit in the police station rather than chasing your target all over town trying to maintain a MITM attack against his cell phone.

Comment Re:How does stingray connect to the wider network? (Score 4, Informative) 90

It doesn't. It just acts as a fake base station; if you happen to connect to one you'll have no service. They don't use these things to intercept your traffic, they can do that Verizon/AT&T/Sprint/T-Mobile's switch, without having to follow you all over town. These devices are used for two purposes:

1. To localize idle cell phones with greater precision than the macro cellular network can.
2. To determine which cell phones are being carried in a specific area.

#2 sounds Orwellian but it has legitimate purposes during criminal investigations, i.e., trying to figure out the IEMIs of burner phones being carried by suspects you have under surveilliance. Once you have the IEMIs you can wiretap them with lawful interception technology built into the phone company's switch.

Comment Re:Ah, come one, don't we trust the Feds? (Score 1) 90

The latest FCC actions were DECADES in the making.

Using a law that was first written when your telephone had a hand crank and last updated when 33.6kbit/s voiceband modems represented the "bleeding edge" of consumer internet connectivity.

This op-ed raises an interesting question: "The real issue is who pays for new Internet investment. Do big users like Netflix and Facebook bear some costs or are these left to the ISPs -- which shift them to the monthly bills of households? For example: In 2014, Netflix agreed to pay Comcast for smoother streaming of its videos. The open question is whether the FCC will permit these interconnection payments and, if so, at what level. But the FCC has weakened the ISPs' bargaining position by requiring them to accept all comers."

Comment Re:Ah, come one, don't we trust the Feds? (Score 1) 90

If we trust FCC to ensure "fairness" of Internet Service Provision:

If the Federal Government can't determine what's fair, then who can?

why don't we trust the Marshals Service to be fair as well? Are they being controlled by a different President or something?

People here are kind of like John Kerry, they were in favor of the Government before they were against it.

Comment Re:I'm dying of curiousity (Score 4, Interesting) 188

If, and thats a big if, VMWare have done anything in violation of the license at all - the "technical FAQ" is very light indeed on actual technical details, instead mostly talking about how Conservancy went to great lengths to do anything they could to open a dialog before suing. The "technical" part of the FAQ is a single diagram which explains very little in how they think VMWares approach is violating either the GPL or copyright.

This is going to be something to watch, as its going to be an interesting one.

Comment Re:If "yes," then it's not self-driving (Score 1) 362

It's worth noting that there is one piece of automation in cars already that does give a different kind of driving license in a lot of places: automatic gear change. If you get a driving license in a car that has an automatic transmission then you can't drive manual cars with it, though the converse is allowed.

And it's silly. You can give an 18yo (around here) that just got his license a Ferrari, that's legal. You can give him a 3500 kg van + 750 kg trailer, that's legal. Of course you shouldn't drive a car you can't handle, but learning it on your own would be no worse than a lot of the other "self-learning" on the road.

Comment Re:Pales to UE4 (Score 1) 74

UE4 is the better engine

Really? Can you provide comparisons?

Actually, I'll answer that for you:

No, you can't, because Source 2 isn't out yet.

If you're comparing UE4 with Source 1, I'd like to point out that while Source has been updated over the years, its core technology is still a decade behind UE4's.

Well, at least it is exciting since maybe we will get a good Valve game out of it. Valve games tend to be defined by their physics puzzles and/or new gameplay innovations. When the technology for having portals in the engine was invented/developed (by others), BAM, we got 2 portal games. A new engine may just bring enough new possibilities that they make some interesting games themselves.

Comment Re:If "yes," then it's not self-driving (Score 1) 362

Even if you can account for such things, how will your autonomous vehicle handle malfunctioning sensors? Aerospace has been working at this for decades and still hasn't figured it all out.

The main reason to have pilots is that you have someone with "skin in the game", not because they're actually good backups. Like in your linked case there's several major pilot errors that were only possible because the safety systems were disabled due to a 30 second glitch in the sensor. After the sensor recovered the pilots were given multiple warnings about what was happening but instead caused such a massive stall that the computer refused to believe the sensors, going silent as the pilots slammed the planed into the ocean killing all on board.

If the computer had taken a HAL 9000 with "I can't let you do that, Dave" and taken the plane out of the stall once it recovered they'd be alive. If the computer had been forced to carry on despite the faulty sensor, it would still have engine power and altitude to infer that air speed is wrong and keep the plane flying and it would almost certainly have done a better job. They died because the default was in any out of the ordinary operation to let the humans take over. It's a better poster child for a self-flying plane than against it. But since the pilots paid with their own lives they become the lightning rod for the anger, while a self-flying plane crashing would be become a corporate nightmare.

Slashdot Top Deals

System going down in 5 minutes.

Working...